Understanding the function of the membrane transporter ABCG2 by comparison with-P glycoproteininteraction with antitumorals, antibiotics, hormones and other compounds

Resumen

Breast Cancer Resistance Protein (BCRP, ABCG2) and P-glycoprotein (P-gp, ABCB1, MDR1) are members of the G and B subfamily of the ATP-binding cassette (ABC) efflux transporters, respectively. Both transporters use the energy of ATP hydrolysis for substrate efflux. ABCG2 was identified in 1998 (Doyle et al., 1998) while P-gp was discovered 20 years earlier (Juliano and Ling, 1976). ABCG2 and P-gp are localized in normal tissues and barriers as well as in tumor cells where they transport a wide variety of compounds, including antitumorals, antibiotics and hormones. Moreover, they show overlapping substrate specificity. These transporters are considered as clinically relevant in drug absorption in the intestine, distribution (e.g., at blood-brain barrier) and elimination in the liver and the kidney; ABCG2, also in drug secretion in the mammary gland (Giacomini et al., 2010). Both are the clinically most important efflux transporters at the blood-brain barrier and the intestinal barrier. Furthermore, they are involved in the multidrug resistance phenomenon (MDR). Despite similar functions, the two proteins show practically no protein sequence identity in the transmembrane domains (TMDs) and only little sequence identity in the nucleotide binding domains (NBDs) (~ 20%) (Li et al., 2007). The present knowledge on P-gp and ABCG2 have been extensively reviewed (e.g. Krishnamurthy and Schuetz, 2006; Sharom, 2008; Robey et al., 2009; Poguntke et al., 2010; Mo and Zhang, 2012). It is thus of great importance in pharmaceutical chemistry and medicine to understand the function of these transporters and to be able to predict their function. In the case of P-gp, broad knowledge has been gained on its biochemistry, structure and function. It has long been known that P-gp binds its substrates in the lipid membrane; as a consequence, substrate binding to P-gp is a two-step process, a lipid-water partitioning step and a transporter binding step in the lipid membrane. In contrast, comparatively, little is known on ABCG2 structural function (e.g. Matsson et al., 2009; Ni et al., 2010). Numerous computational approaches or models based on Quantitative Structure-Activity Relationships (QSAR) models, Structure–Activity Relationships (SARs) analysis, pharmacophore modeling and molecular docking were developed to predict P-gp inhibitors or substrates (Chen et al., 2012); but only limited models can give satisfactory predictions. In the case of ABCG2, the number of approaches is lower. However, despite different approaches, the nature of drugtransporter interactions in the case of ABCG2 is not yet very clear. A model that can predict ABCG2 substrates and inhibitors accurately is still lacking. In addition, no direct molecular model explaining how ABCG2 works has been reported so far. The main aim of the present memory was to understand the ABCG2 function by comparison witn P-gp. For this goal, several specific aims were proposed. The first specific aim of the present investigation was to gain further insight into the similarities and differences between ATPase kinetics and the substrate binding of ABCG2 and P-gp; and in addition, to attempt a prediction explaining the substrate interactions of compounds at the molecular level with both transporters. For this purpose we chose 28 compounds, including electrically neutral, cationic, anionic and zwitterionic species, 20 of which interacted with both transporters.Introduction 4 As it has been already mentioned, one of the major constituents of ABCG2 and P-gp substrates are hydrophobic chemotherapeutics, including topoisomerase inhibitors, anthracyclines, camptothecin analogs, tyrosine kinase inhibitors, antimetabolites, anthracyclines, vinca alkaloids and taxanes (Duan et al., 2009; Mo and Zhang, 2012). Despite the fact that the interaction of these antitumorals with ABCG2 and P-gp have been investigated by means of in vitro cell-based and ATPase activity assays as well as by means of in vivo experiments, a general understanding regarding their interaction with these two ABC transporters at the molecular level remains to be clarified. Therefore, several antitumorals were also included in the list of compounds to study and explain the substrate interaction with P-gp and ABCG2. In the field of ABC transporter analysis, a certain disparity has been developed in data interpretation between groups focus in protein science and groups focus in membrane science. However, both items (protein and membrane) function synergistically. Most drugs reach their target by passive diffusion across several membrane lipid bilayer barriers. Transporters such as ABCG2 and P-gp compete with passive drug influx into the cell by binding drugs in the lipid bilayer membrane and moving them back to the extracellular environment at expenses of ATP hydrolysis. To assess whether a compound is a substrate or an inhibitor of a particular transporter the International Transporter Consortium (Giacomini et al., 2010) recommends bi-directional transport assays with polarized confluent cell monolayers expressing the transporter of interest. Measurement of the substrate-induced ATP hydrolysis was in contrast not recommended because of inconsistency between the ATPase activity and the transport rate of some substrates and inhibitors and a high incidence of false positive and false negatives. However, the rate of ATP hydrolysis and the rate of substrate transport unambiguously correlate, provided, the passive flux across the bilayer membrane is taken into account in a quantitative manner (Seelig, 2007). The second specific aim of the present study was to test the correlation between ATPase activity of both transporters and the rate of substrate transport in transport and accumulation assays by taking into account that most drugs can cross the bilayer membrane by passive diffusion. For this analysis we chose steroid hormones. The interaction between steroid hormones and the two ABC transporters, P-gp and ABCG2, has been extensively studied and published and we used the large amount of transport and accumulation assays provided in literature to compare them with our ATPase activity and SAM results. Steroid hormones moreover have the advantage to be electrically neutral which allows ignoring charge effects and make easier the explanations of the interactions between transporters and steroids. Another group of compounds with great scientific interest that interact with ABC transporters is the antibiotics. Antibiotics for the treatment of infections include fluoroquinolones, a class of synthetic antimicrobial with remarkably broad spectrum and high antibacterial potency. In a similar way as the second purpose, the third specific aim of this investigation was to correlate ABCG2 and P-gp- ATPase activity obtained for five common fluoroquinolones (ciprofloxacin, enrofloxacin, moxifloxacin, Introduction 5 norfloxacin, pefloxacin) with obtained and published data from accumulation and transport cell-based assays, which allows explaining the interactions between these fluoroquinolones and ABCG2 and P-gp at the molecular level. Additionally, to complete the characterization of the interaction of these fluoroquinolones with ABCG2, we evaluate the in vivo interaction of moxifloxacin, norfloxacin and pefloxacin with ABCG2.